Your browser doesn't support javascript.
loading
Meta-learning based blind image super-resolution approach to different degradations.
Yang, Zhixiong; Xia, Jingyuan; Li, Shengxi; Liu, Wende; Zhi, Shuaifeng; Zhang, Shuanghui; Liu, Li; Fu, Yaowen; Gündüz, Deniz.
  • Yang Z; College of Electronic Engineering, National University of Defense Technology, Changsha, 410073, China. Electronic address: yzx21@nudt.edu.cn.
  • Xia J; College of Electronic Engineering, National University of Defense Technology, Changsha, 410073, China. Electronic address: j.xia10@nudt.edu.cn.
  • Li S; College of Electronic Engineering, Beihang University, Beijing, 100191, China. Electronic address: LiShengxi@buaa.edu.cn.
  • Liu W; College of Electronic Engineering, National University of Defense Technology, Changsha, 410073, China. Electronic address: liuwende@nudt.edu.cn.
  • Zhi S; College of Electronic Engineering, National University of Defense Technology, Changsha, 410073, China. Electronic address: zhishuaifeng11@nudt.edu.cn.
  • Zhang S; College of Electronic Engineering, National University of Defense Technology, Changsha, 410073, China. Electronic address: zhangshuanghui@nudt.edu.cn.
  • Liu L; College of Electronic Engineering, National University of Defense Technology, Changsha, 410073, China. Electronic address: li.liu@oulu.fi.
  • Fu Y; College of Electronic Engineering, National University of Defense Technology, Changsha, 410073, China. Electronic address: fuyaowen@nudt.edu.cn.
  • Gündüz D; Department of Electrical and Electronic Engineering, Imperial College London, London, SW72AZ, UK. Electronic address: d.gunduz@imperial.ac.uk.
Neural Netw ; 178: 106429, 2024 Oct.
Article en En | MEDLINE | ID: mdl-38901090
ABSTRACT
Although recent studies on blind single image super-resolution (SISR) have achieved significant success, most of them typically require supervised training on synthetic low resolution (LR)-high resolution (HR) paired images. This leads to re-training necessity for different degradations and restricted applications in real-world scenarios with unfavorable inputs. In this paper, we propose an unsupervised blind SISR method with input underlying different degradations, named different degradations blind super-resolution (DDSR). It formulates a Gaussian modeling on blur degradation and employs a meta-learning framework for solving different image degradations. Specifically, a neural network-based kernel generator is optimized by learning from random kernel samples, referred to as random kernel learning. This operation provides effective initialization for blur degradation optimization. At the same time, a meta-learning framework is proposed to resolve multiple degradation modelings on the basis of alternative optimization between blur degradation and image restoration, respectively. Differing from the pre-trained deep-learning methods, the proposed DDSR is implemented in a plug-and-play manner, and is capable of restoring HR image from unfavorable LR input with degradations such as partial coverage, noise addition, and darkening. Extensive simulations illustrate the superior performance of the proposed DDSR approach compared to the state-of-the-arts on public datasets with comparable memory load and time consumption, yet exhibiting better application flexibility and convenience, and significantly better generalization ability towards multiple degradations. Our code is available at https//github.com/XYLGroup/DDSR.
Asunto(s)
Palabras clave

Texto completo: 1 Banco de datos: MEDLINE Asunto principal: Redes Neurales de la Computación Límite: Humans Idioma: En Año: 2024 Tipo del documento: Article

Texto completo: 1 Banco de datos: MEDLINE Asunto principal: Redes Neurales de la Computación Límite: Humans Idioma: En Año: 2024 Tipo del documento: Article